An efficient image steganography approach based on QR factorization and singular value decomposition in non‐subsampled contourlet transform domain

• Published in: Security and privacy 2022-07, Vol.5 (4), p.n/a
• Main Authors: Yasmeen, Fauzia, Uddin, Mohammad Shorif
• Format: Article
• Language: English
• Subjects: imperceptibility; non‐sub‐sampled contourlet transform; QR factorization; robustness; singular value decomposition; steganography
• ISSN: 2475-6725; 2475-6725
• DOI: 10.1002/spy2.229

As the amount of information communicated through the virtual world grows dramatically day by day, data security becomes a significant concern while communicating over the internet. Steganography can play an integral part in safeguarding the data from unauthorized users through a hiding mechanism. This article describes a three‐step steganographic algorithm. This algorithm applied QR factorization and singular value decomposition (SVD) in the non‐subsampled contourlet transform (NSCT) domain. First, this algorithm scrambles the secret image using Arnold transforms, and then NSCT decomposes the carrier and the scrambled hidden image into coefficient sub‐bands. Second, the QR factorization and SVD are used on the specific coefficients of carrier and scrambled secret images, respectively. Finally, the modified secret image inserts into the carrier image for communication. The reverse mechanism is applied to discover the hidden message at the receiving end. The methodology gives superior imperceptibility as well as robustness. Furthermore, the anticipated strategy is compared with current methods in terms of effectiveness.